Opinion spread

This is a simple model of how an opinion spreads through a community. Each individual has a number of opinions as a list of integers. They can change their opinion by changing the numbers in the list.

Agents can change their opinion at each step. They choose one of their neighbors randomly, and adopt one of the neighbor's opinion. They are more likely to adopt their neighbors opinion if the share more opinions with each other.

using Agents
using AgentsPlots
using Random

Building the model

1. Model creation

mutable struct Citizen <: AbstractAgent
    id::Int
    pos::Tuple{Int,Int}
    stabilized::Bool
    opinion::Array{Int,1}
    prev_opinion::Array{Int,1}
end

function create_model(; dims = (10, 10), nopinions = 3, levels_per_opinion = 4)
    space = GridSpace(dims, periodic = true, moore = true)
    properties = Dict(:nopinions => nopinions)
    model = AgentBasedModel(
        Citizen,
        space,
        scheduler = random_activation,
        properties = properties,
    )
    for cell in 1:nv(model)
        add_agent!(
            vertex2coord(cell,model),
            model,
            false,
            rand(1:levels_per_opinion, nopinions),
            rand(1:levels_per_opinion, nopinions),
        )
    end
    return model
end

create_model (generic function with 1 method)

2. Stepping functions

function adopt!(agent, model)
    neighbor = rand(space_neighbors(agent, model))
    matches = model[neighbor].opinion .== agent.opinion
    nmatches = count(matches)
    if nmatches < model.nopinions && rand() < nmatches / model.nopinions
        switchId = rand(findall(x -> x == false, matches))
        agent.opinion[switchId] = model[neighbor].opinion[switchId]
    end
end

function update_prev_opinion!(agent, model)
    for i in 1:(model.nopinions)
        agent.prev_opinion[i] = agent.opinion[i]
    end
end

function is_stabilized!(agent, model)
    if agent.prev_opinion == agent.opinion
        agent.stabilized = true
    else
        agent.stabilized = false
    end
end

function agent_step!(agent, model)
    update_prev_opinion!(agent, model)
    adopt!(agent, model)
    is_stabilized!(agent, model)
end

agent_step! (generic function with 1 method)

Running the model

levels_per_opinion = 4
model = create_model(nopinions = 3, levels_per_opinion = levels_per_opinion)

"run the model until all agents stabilize"
rununtil(model, s) =
    count(getproperty.(values(model.agents), :stabilized)) == prod(model.space.dimensions)

agentdata, _ = run!(model, agent_step!, dummystep, rununtil, adata = [(:stabilized, count)])

(250×2 DataFrame
│ Row │ step  │ count_stabilized │
│     │ Int64 │ Int64            │
├─────┼───────┼──────────────────┤
│ 1   │ 0     │ 0                │
│ 2   │ 1     │ 83               │
│ 3   │ 2     │ 68               │
│ 4   │ 3     │ 75               │
│ 5   │ 4     │ 80               │
│ 6   │ 5     │ 70               │
│ 7   │ 6     │ 74               │
⋮
│ 243 │ 242   │ 97               │
│ 244 │ 243   │ 95               │
│ 245 │ 244   │ 98               │
│ 246 │ 245   │ 98               │
│ 247 │ 246   │ 99               │
│ 248 │ 247   │ 98               │
│ 249 │ 248   │ 95               │
│ 250 │ 249   │ 100              │, 0×0 DataFrame
)

Plotting

The plot shows the number of stable agents, that is, number of agents whose opinions don't change from one step to the next. Note that the number of stable agents can fluctuate before the final convergence.

plot(
    1:size(agentdata, 1),
    agentdata.count_stabilized,
    legend = false,
    xlabel = "generation",
    ylabel = "# of stabilized agents",
)

Animation

Here is an animation that shows change of agent opinions over time. The first three opinions of an agent determines its color in RGB.

levels_per_opinion = 3
ac(agent) = RGB((agent.opinion[1:3] ./ levels_per_opinion)...)
model = create_model(nopinions = 3, levels_per_opinion = levels_per_opinion)
anim = @animate for sp in 1:500
    step!(model, agent_step!)
    p = plotabm(model, ac = ac, as = 12, am = :square)
    title!(p, "Step $(sp)")
    if rununtil(model, 1)
        break
    end
end
gif(anim, "opinion.gif")